Re: LD not precious?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 01/14/2010 06:38 PM, Eric Blake wrote:
> According to Philip A. Prindeville on 1/14/2010 4:39 PM:
>>> And the *_TOOL will then cause the right thing to happen if you
>>>   ./configure --host=foo-bar-alias
>>>
>>> (i.e., try foo-bar-alias-ld first, just like with $CC).
>>>
>>> Cheers,
>>> Ralf
>>
>> Would be nice if c.m4 contained that...
> 
> Why?  The whole point is that you can compile and link a program using
> just $CC, so LD should not be needed in those cases.  It is only desirable
> to mark it precious in the cases where LD will be used in addition to CC,
> but since that should not be the default for all packages, then the
> packages that WANT to use LD should be the ones responsible for marking it
> precious.  I don't see anything wrong with autoconf's current behavior.

Right.  And I'm saying that in 99% of the cases where cross-compilation is happening (i.e. $host != $target), that the loader *is* needed.

Regrettably, most people do an extremely bad job writing cross-compilation friendly packages.

Good tools could go a long way helping here.


_______________________________________________
Autoconf mailing list
Autoconf@xxxxxxx
http://lists.gnu.org/mailman/listinfo/autoconf

[Index of Archives]     [GCC Help]     [Kernel Discussion]     [RPM Discussion]     [Red Hat Development]     [Yosemite News]     [Linux USB]     [Samba]

  Powered by Linux